Electrochemical capacitors (referred to as ultracapacitors or supercapacitors) are devices that use electrolytes that contain and conduct ions in conjunction with electrodes that are ionically and electronically conductive, as a system to store energy. Unlike batteries, which store charge within the bulk electrode material itself, ECs store charge at or near the interface between the electrolyte and electrode material making EC storage a surface phenomenon. EC devices store charge via one of two mechanisms: electric “double-layer” or faradaic “pseudocapacitance”. Alternatively, intercalation materials typical of secondary lithium ion batteries can be combined with materials exhibiting one or more of these aforementioned storage mechanisms as a way to increase the amount of energy stored.
Electric double layer capacitors (EDLCs) store charge at the interface between the ion-rich electrolyte and an electronically conductive electrode comprising material such as activated carbon (AC). The amount of charge stored is a function of this interfacial area, which is in turn related to the electrode surface area accessible by the electrolyte-solvated ions. AC electrodes typically have surface areas between 1000 and 3000 m2/g achieved by a high concentration of micropores (pore diameter<2 nm). This concentration of micropores paradoxically serves to limit the capacitance to a range between 15 and 30 uF/cm2 because micropores are too small to accommodate the solvated ions necessary for the double layer mechanism. AC electrodes are typically fabricated as films formed from a paste comprising powdered AC, binder and conductivity-enhancing carbon. While the accessibility of the electrolyte to much of the surface area of such an electrode is limited, thus limiting the electrode capacitance and ultimately device energy density, the micropore geometries also serve to limit electrode charge/discharge rates and ultimately limit device power density.
Pseudocapacitance involves electron transfer in oxidation/reduction reactions that take place between the electrolyte ions and the electrode active materials over a voltage range similar to that of the EDLC. Electrode active materials being researched for this type of behavior include graphitic carbons, conductive polymers and transition metal oxides.
Transition metal oxides exhibit characteristics both desirable and otherwise for EC devices. In general, these oxides exhibit improved specific capacitance vs. carbon materials, but they typically do so at a monetary cost or at the expense of cycle life and power density. As an example of the former, the specific capacitance of hydrous ruthenium oxide has been demonstrated to be higher than 700 F/g, which is far greater than any known carbon EDLC. However, the cost and lack of abundant supply of the ruthenium materials prohibit broad commercial use.
Lower cost, plentiful transition metal oxides such as manganese oxide, nickel oxide and others have been investigated and in some cases commercialized. The specific capacitance of these materials ranges from approximately 200 F/g for powder/paste derived thick films (tens of microns to hundreds of microns) to more than 500 F/g for very thin planar films of less than 100 nanometers. This difference in capacitance (and ultimately energy density) demonstrates the surface nature of the pseudocapacitance charge storage mechanism. However, power density also is limited by the surface nature of pseudocapacitance as the longer ion diffusion length of thicker material serves to reduce reaction rate. The power density of electrodes using these oxides is further limited by the very low intrinsic electronic conductivity of these materials. These oxides also exhibit limited cycle stability and operating voltage ranges limited to approximately 0.8 volts.
Metal oxide paste-derived electrodes combine the powdered metal oxide with a binder and conductivity-enhancing carbon similar to the EDLC activated carbon electrode. The resulting electrode is limited by the characteristics noted above and also the lack of surface area readily accessible to the electrolyte. The limited pore volume also impedes ion flow to the inner electrode surfaces further limiting the pseudocapacitance reaction rate.
EC devices employing a metal oxide thick film configure their systems in such a way so as to use the oxide-based electrode as a largely static energy storage element creating an offset voltage from which a carbon electrode only is cycled through charge and discharge. This configuration is known in the art as an asymmetric EC.
As a way of using pseudocapacitance to increase device energy, U.S. Pat. No. 6,339,528 discloses the synthesis of an layered, amorphous phase manganese oxide containing potassium ions on a carbon powder, which is then ground to form a paste used with a binder to form an electrode for an aqueous-only ultracapacitor cell. Others have suggested similarly coating carbon nanotubes with amorphous manganese oxide subsequently mixing the coated nanotubes with a binder to form an EC electrode. While each of these approaches offer improved rate performance resulting from the reduced ion diffusion length vs. the simple oxide paste electrodes, the increase in stored energy is limited by the pseudocapacitance charge storage mechanism provided by the amorphous oxides used.
Long et al. (see for example, 20080247118; 20080248192; and 20100176767) have proposed an approach for addressing these shortcomings by applying a very thin coating of poorly crystalline MnO2 or iron oxide to a carbon structure. In doing so, the high capacitance and fast reaction rate of the thin film approach is preserved. Further, the 3-dimensional carbon structure provides a low (electronic) resistance path to the current collector and an open porosity providing much improved electrolyte ion access to the MnO2 or iron oxide active material. The synthesis approach suggested by Long involves the reduction of permanganate or potassium ferrate(VI) on the surface of the carbon. This takes place as the coating is deposited utilizing the carbon as a sacrificial reductant to synthesize the oxide. The oxide deposition method suggested by Long results in a conformal coating of the carbon structure. Again, the oxides formed in these cases are of the amorphous variety, providing pseudocapacitive behavior which produces less energy storage capacity than intercalation materials typically provide.
While the previously discussed improvements in EC technology are highly significant, there remains a need in the art for EC devices and therefore EC electrodes having expanded operating voltage range and increased the storage capacity while also exhibiting high power density.
Lithium ion secondary batteries operate through the intercalation/de-intercalation of lithium ions into and out of the solid bulk electrode materials. Today's lithium ion electrode materials typically comprise a graphite-based anode and a transition metal oxide (typically cobalt, nickel or manganese) cathode. During the charging cycle, electrons are removed from the cathode, which causes charge-compensating lithium ions to be released into and dissolved in the electrolyte where they migrate towards the anode; while electrons are simultaneously added to the anode causing lithium ions to be inserted into the anode. The opposite occurs during discharge.
Ion insertion (diffusion) into the bulk oxide cathode material takes place in vacancies present in the oxides. The rate of this process is affected by the diffusion length and the accessibility of the electrolyte to the oxide. This rate in turn affects the instantaneous power capability of the battery.
The fabrication method of these electrodes relies on powdered active materials formed into a paste including binder material and (electronic) conductivity enhancing carbon. The thickness of these cathodes ranges from 30 micrometers for high power (low energy) batteries to 200 micrometers for high-energy (low power) versions. Typical oxide powder particle sizes vary from hundreds of nanometers to a few micrometers in diameter. Lithium ions diffuse as much as a few microns into the bulk oxide particle. The charge-compensating electrons from the oxide must then traverse the low-conductivity oxide and electrolyte to complete the circuit.
Ion diffusion into the solid-state electrode particles induces mechanical stress on the oxide crystal lattice as it expands to accommodate the ion insertion. These expansion/contraction cycles cause the eventual breakdown of the oxide limiting device cycle life. It is therefore preferable for the diffusion into the oxide to be as shallow as possible and to choose ion/oxide systems that exhibit minimal expansion.
While the previously discussed improvements in secondary battery technology are highly significant, there remains a need in the art for secondary battery devices and therefore secondary battery electrodes having improved cycle life, shorter recharge time and generally increased usable storage capacity at elevated power levels.